^{page 151 note 2} ‘… propterea quod a strepitu seculari remotior, bibliotecae admodum videtur conveniens’: Ep. Ac. Ox. i, 246.

^{page 151 note 3} One of the very few changes is the disappearance of the small bellcote on the roof at the south-west corner of Anstey Duke Humphrey's library. This must have been added to it by Bodley to take his library bell which he gave in 1611. Since the recent reconstruction the bell, which had been mounted rather incongruously on the floor of Selden End, has again been hung in this position, but inside the roof instead of above it, so that no bellcote is required.

^{page 151 note 4} The earliest stages of the building can be traced in the university's begging letters between 1423 and 1427: Ep. Ac. Ox. i, 10. In 1423 the bishop of Bath and Wells is told of ‘quodam scolarum edificio jam incepto’: in 1426 ‘scolarum nostrarum … inchoata fundamenta’ are mentioned to the archbishop of Canterbury: by 1427 ? it was possible to speak to the master of St. Thomas Hospital, London, of ‘muri scolarum predictarum jam terre suri, gentes de sinu’.

^{page 152 note 1} The most useful earlier attempt to do this is in Hope's, W. St. John article on ‘The Heraldry and Sculpture of the Vault of the Divinity School’ (Arch. Journ. lxxi (1914), 217—60Google Scholar), but further sources have been printed since he wrote; the same is true of the historical summary in V.C.H. Oxon. iii (1954), 44–45, and of the architectural description in R.C.H.M. Oxford (1939), pp. 5–10, both of which contain valuable information.

^{page 152 note 2} The details concerning these appointments are summarized in V.C.H. Oxon. iii (1954), 44: for the careers of these and other Oxford masons mentioned in this paper see Gee, E. A., ‘Oxford Masons 1370–1530’ in Arch.Journ. cix (1952), 54–131Google Scholar.

^{page 152 note 3} ‘… dicta Universitas vult quod dictus Thomas retrahet… supervacuam talem curiositatem … in tabernaculis ymaginum, caesimentis, et fylettis et in aliis frivolis curiositatibus que ad rem non pertinent sed ad nimias et sumptuosas expensas … et ad nimiam dicti operis tardacionem.’ It is interesting to note that this change in architectural direction was attributed by the university to the views of ‘plures magnati regni et alii sapientes’, who were moved by considerations of taste as well as of time and money (Ep. Ac. Ox. i, 192).

^{page 152 note 4} This is the point to which the panelling of the northern buttresses also extends (pl. xxix a and fig. 1), and it may well represent the intended wallplate level of the original building. If so, the vault could have sprung from a point within the maximum depth of the buttresses, and below the heads of the main windows and could have been comfortably contained within a roof of quite low pitch. The diagonal line of decoration on Winchcombe's buttresses (fig. 1) seems intelligible only as an indication of the line of stress from a vault springing from this point.

^{page 152 note 5} Both the east and west doors are decorated in Winchnaculis combe's manner and there is evidence that both ends were panelled, at least to some extent, as were his northern buttresses. See pis. xxviii, xxxviii b, and xxxvi.

^{page 153 note 1} As shown on John Bereblock's drawing of 1566: see pl. XLVII a.

^{page 153 note 2} His will was proved in the Chancellor's Court on 6th October 1449: Reg. Canc., i, 191–2.

^{page 153 note 3} ‘… ad supervidendum et indicandum altitudinem dicte fabrice novarum scolarum’: see the accounts of W. Churche, supervisor of the works, printed and discussed by Cox, A. D. M. in Oxoniensia, xxi (1956), 48–60Google Scholar.

All the information here given for work done in 1452–3 is derived from this very informative document which was not available to earlier writers on this subject.

^{page 153 note 4} A total of 109 feet of stone for the ‘corbell tabull’ is recorded which would be sufficient for the whole of one side, presumably the north, and a large part of one end of the building.

^{page 154 note 1} Four nobles were taken from the Proctor's Chest ‘pro coopertura fabricae novarum scolarum’: Reg. Aa. f. 107r.

^{page 154 note 2} Sixteen pence were paid ‘pro stramine ad cooperturam novarum scolarum’ and 3s. 6d. for labour: Med. Arch. Univ. Oxon. ii, 295.

^{page 154 note 3} John Godard of Bucklebury, Berks., binds himself to make by St. Andrew's Day ‘XXXVII ambones cum scabellis correspondentibus pro novis scolis sacre theologie’: Reg. Canc., ii, 185.

^{page 154 note 4} ‘pro inscisione capitum ambonum in novis scolis iii lib. vi s. viii d.’, Med. Arch. Univ. Oxon. ii, 299.

^{page 154 note 5} £10 is paid to the supervisors of the new schools ‘pro completione edificationis muri’. Med. Arch. Univ. Oxon. ii, 299.

^{page 154 note 6} ‘priusquam nova fit libraria locum aptum non habemus’: Ep. Ac. Ox. ii. 391.

^{page 154 note 7} Forty shillings was received from the Chest of Five Keys ‘pro reparatione facta circa tectum novarum scolarum’, and expenditure included three shillings ‘pro clausura fenestrarum superiorum et obstruccione foraminum’, and twenty-six shillings and eight pence ‘protectura novarum scolarum theologie et pro renovatione plumbi eiusdem tecture’: Med. Arch. Univ. Oxon. ii, 306–8. The use of the words ‘tectum’ and ‘tectura’ for this roof suggests that it was something more permanent than the thatched ‘coopertura’ of 1464/5, which may have been only a temporary covering carried on scaffold poles to enable work to proceed during wet weather, much as a tarpaulin might be used today. The ‘foramina’ filled in 1472/3 could have been holes left in the walls to support scaffolding.

^{page 154 note 8} Med. Arch. Univ. Oxon. ii, 291 and 324.

^{page 155 note 1} I am greatly indebted to Mr. J. H. Harvey, F.S.A., for checking for me the exact wording of the passage on the Divinity School in the Cambridge manuscript of William Worcester's Itineraries (Corpus Christi College MS. 210) transcribed for his forthcoming edition of that work. William was at Oxford, staying at the Bull, from 11th to 19th August 1480, and he writes thus of the Divinity School (p. 209): ‘Scola Theologie nova cum libraria desuper … continet in longitudine . 30 . virgas quelibet virga . 3 . pedum et [blank] Et in latitudine continet . 32. pedes et in altitudine a fundo usque ad superiorem walplate de frestone . 80 . pedes: et in altitudine ad voltam novam sive arcum modo operatum in anno Christi. 1480 . a terra sive piano solo continet. 32. pedes.’

^{page 155 note 2} It is clear that the distinction made between the internal height of the vault ‘a terra’ and the overall external height ‘a fundo’ is deliberate and implies that the depth of the foundation is included in the latter figure. Mr. Harvey has suggested to me that the miscalculation of the overall height may be partly due to William being given the height to the apex of the gable rather than to the top of the walls. Even so the error must be considerable, for it could hardly have been more than 70 feet from the foundation to the gable end: see pl. xxviii.

^{page 156 note 1} In that year the bishop of London was told ‘Ipsi vero omnes libri, quum primum ad nos dabuntur, una cum extenaliis libris qui in antiqua libraria Universitatis adhuc sunt, in tuam novam librariam cathenandi citissime transferentur.’ This passage shows that the bishop, whose benefaction had paid for the vault of the Divinity School, was thereby regarded as essentially the donor of the library also: Ep. Ac. Ox. 545.

^{page 156 note 2} In 1488 a donor of thirty-one books is told that his volumes ‘sunt firmissime cathenati in nova nostre Universitatis libraria: que eos maxima quadam gratulatione propterea suscepit quod sunt primitie omnium librorum quos consusceptura est’: Ep. Ac. Ox. 545.

^{page 156 note 3} For the history of Oxford libraries at this time see J. N. L. Myres in The English Library before 1700, ed. F. Wormald and C. E. Wright, 1958, pp. 238–43, and the references there given.

^{page 156 note 4} Calendar of State Papers Domestic 1598–1601, p. 35.

^{page 156 note 5} W. D. Macray, Annals of the Bodleian Library (ed. 1890), pp. 168–9.

^{page 156 note 6} The existence of Wren's ‘ramping arches’ was verified in 1922 in an excavation (in which I took part as a student) conducted by the late L. H. D. Buxton, F.S.A. The masonry of the additional buttresses had become extensively decayed before the recent restoration (pl. xxx a). Since they would have been very expensive to renew, and now served no structural purpose, they were removed. Compare pls. xxx a and XLVII b for the appearance of the south wall from Exeter College garden before and after restoration.

^{page 156 note 7} That the library floor was originally laid directly on the vault is shown by the little flat arches visible on pl. xxxiii that provided a level surface on which to lay the main floor timbers. This was quite safe so long as the library consusceptura tents were limited to a few hundred manuscripts on lectern desks. No one in 1480 could anticipate that the vault would one day be expected to carry the weight of Bodley's tall bookcases and thousands of printed books. It was probably this change of load more than any other single factor that produced the instability which Wren was called on to correct. It is a remarkable tribute to Orchard's architectural skill that his vault sustained this unintended weight so long.

^{page 158 note 1} All the drawings, including that for pl. xxvm, are by Mr. Potter to whom I am deeply indebted for much help in preparing this paper.

^{page 158 note 2} The evidence on this last head will not be discussed in this paper. It amounts in brief to the fact, whose implications were not perhaps fully realized before, that the present wooden fittings of Selden End with its galleries mask several original features of the room as first designed. It has always been realized, for example, that the original plan provided for lighting by three great windows on the west elevation. Only the central one is now in use. The outlines of the others are still visible externally but they were entirely covered internally by the gallery and wall-shelving. Their internal recesses, however, have now been ingeniously used by Mr. Potter to provide new staircases to the western sections of the gallery (pl. xxxviii a) previously only accessible by unsightly bridges, now removed, spanning the north and south windows. How these sections of the gallery were originally reached remains something of a mystery, Another feature at first intended to be seen but later encased in the panelling is the great stone arch with its fine mouldings separating Selden End from Duke Humphrey's library (pl. xxxvii).

^{page 159 note 1} In his will Bodley expressed the hope that it would be possible ‘to make the ascent more easye and graceful …’.

^{page 159 note 2} This porch, presumably similar to that shown by Bereblock at the west end, has left the scar of its abutment showing above and around the door which it covered, Part of its projecting foundation was found under the Proscholium pavement during the recent excavations.

^{page 160 note 1} Unless otherwise stated, the photographs are by Mr. J. W. Thomas (Thomas Photos.) taken on behalf of the Bodleian Library. His skill and interest have provided a splendid pictorial record of the whole reconstruction.

^{page 161 note 1} As, e.g., at New College, Winchester, Balliol, and elsewhere.

^{page 161 note 2} For the phases through which perpendicular architecture passed in Oxford during the fifteenth century see the stimulating article by Davis, R. H. C. ih Oxoniensia, xi/xii (1946/7), 75–89Google Scholar. I borrow from him the term ‘drop tracery’ for the type in which the heads of windows, generally four-centred, are filled with rectilinear tracery from well below the spring of the arch.

^{page 162 note 1} The seating for a different set of mouldings can be seen above several of the corbels on pls. XLI and XLII, especially XLI b and XLII d.

^{page 162 note 2} Oxoniensia, xxi (1956), 58Google Scholar.

^{page 163 note 1} Smith, A.H., New College, Oxford, and its Buildings (1952), pp. 53–54Google Scholar. Here too the lecterns were 5 ft. 6 in. high.

^{page 163 note 2} In 1489 the bishop of Lincoln wrote to suggest that these old fittings, ‘illas quidem sediculas cum antepositis receptaculis librorum quibus in veteri biblioteca Universitatis supra domum congregationis omnesjam diu usi fueramus’, might be used to furnish the new School of Canon Law. The university replied, with typical bureaucratic caution, that no decision would be taken about their future use which disregarded the bishop's admirable suggestion.

^{page 164 note 1} Letters of Sir Thomas Bodley to Thomas James, ed. G. W. Wheeler (1926), p. 139.

^{page 165 note 1} In addition to these two sets of corbels, totalling ten in all, there are two which are rough, uncarved blocks, evidently later replacements for earlier ones damaged at some stage in repairs to the roof.

^{page 165 note 2} Mr. I. G. Philip, Deputy Keeper of the University Archives, kindly tells me that the University Accounts between 1596 and 1612 contain three items indicating structural expenditure on the roof. These all occur in the years 1597–9 and total £54. 10s. 2½d.

^{page 165 note 3} Calendar of State Papers Domestic 1598–1601, p. 174.

^{page 166 note 1} Oxford University Gazette, 20th March 1877, p. 288.

^{page 166 note 2} A very similar conversion was effected at this time, and for the same reason, on the somewhat similar fifteenth-century roof of the refectory at St. Frideswide's Priory. This was refurnished as a college library for Christ Church in imitation of Bodley's work in Duke Humphrey by Otho Nicholson in 1610–12. Oxoniensia, xxvi/xxvii (1961/2), 215–43Google Scholar.

^{page 166 note 3} Annals ofthe Bodleian Library (ed. 1890), p. 168. The galleries can be seen in section in Wren's drawing, pl. li.

^{page 167 note 1} Mr. J. H. Harvey has reminded me that Chichele was one of the group of ‘prehistoric Wykehamists’, the band of scholars established by William of Wykeham in the Soke at Winchester before the college buildings were ready for their occupation in 1394. He probably was well aware of what was going on thirty years later in providing a library over Fromond's Chantry.

^{page 167 note 2} Money was paid in 1438/9 (custus domorum) ‘Johanni carpentario London laboranti circa facturam deturvalvarum hostij capelle Fromond hostij librarij in eadem capella’. At that date the staircase, too, is mentioned in payments ‘Galfrido plumbario laboranti per i diem supra tectum ascensorii capelle Fromond’. which establish the priority of the library at Winchester over that at Oxford, to the kindness of Mr. J. H. Harvey.

^{page 167 note 3} See the article by the late Herbert Chitty, F.S.A., in Archaeologia, lxxv (1926), 139–58, who does not, however, deal with the library side of the structure, except to note that trouble was experienced with its windows in 1457/8 when five shillings were paid to Stephen Glasyer, ‘emendanti diversos defectus in capella Fromond et in superiore domo eiusdem ordinata pro libraria, destructa et deturvalvarum pata per columbas et alios volucres’.

^{page 167 note 4} Reproduced as the frontispiece to A. H. Smith, New College and its Buildings (1952), and discussed ibid., p. 179.

^{page 167 note 5} Reg. Canc., ii, 185.

^{page 167 note 1} Med. Arch. Univ. Oxon. ii, 324.

^{page 167 note 2} R.C.H.M. Oxford, pp. 6–7.

^{page 167 note 3} It is worth noting that Fromond's Chantry was provided with an elaborate stone vault decorated with a wealth of heraldic and other carved bosses. Herbert Chitty argued (op. cit.) that the carving of these bosses was being done as late as 1445, the year after Chaundler was Proctor. He was no doubt aware of what was going on at Winchester in those years. The Winchester vault may thus have been still in his mind when money at last became available to complete the Divinity School with even greater splendour and an even more elaborate heraldic display than had been devised for Fromond's Chantry over thirty years before.